RECFA Label-Free Intrinsic Imaging- LFII™ Particle Physics and the Future of Biomedicine LFII™ technology in High Performance Capillary Electrophoresis in life sciences, diagnostics and analytical chemistry John Hassard, Department of Physics, Imperial College Founder and CTO deltaDOT www.deltadot.com [email protected]May 11 th 2007
22
Embed
RECFA Label-Free Intrinsic Imaging- LFII™ Particle Physics and the Future of Biomedicine LFII™ technology in High Performance Capillary Electrophoresis.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
RECFA
Label-Free Intrinsic Imaging- LFII™ Particle Physics and the Future of Biomedicine
LFII™ technology in High Performance Capillary Electrophoresisin life sciences, diagnostics and analytical chemistry
John Hassard, Department of Physics, Imperial CollegeFounder and CTO deltaDOT www.deltadot.com [email protected]
• Conventional techniques (CE, MS, HPLC, 2D SDS PAGE…) separate in one or two usually orthogonal dimensions
• Key performance indicators are sensitivity, resolution, quantification accuracy, throughput, dynamic ranges in more than one parameter (eg Mw, concentration)
• Typically, there is a play-off between key parameters. Eg increase in sensitivity can result in decrease in resolving power (more label needed); increase in throughput can lead to less quantification
• LFII is a new approach which rewrites these trade-offs by using proprietary multipixel approach based on HEP algorithms.
Resolution and spectral analyses:
LFII draws on techniques from other disciplines Use spectral information, excellent instrumentation and time-elapse algorithms
Data
Informatio
n
Knowledge
astrophysics
Particle physics
Multipixel Detection and Vertexing
5 positions(214, 254, 280nm)
1:1 image
Standard capillary
512 pixel PDA
UV Light Source
Optics
UV filter
Optics
Detector
Signal
DataProcessing
(GST & EVA)
Separation
512 electropherograms
Generalised Separation Transform
R 1/PTIn a 4 tesla field:
In z-t space in a separation, a similar transform exists in v-space.This allows us to aggregate millions of images from multiple pixels in an unbiased way – the GST
RPair-wise summation of x-y points in 2D space allows a peak in R-space to be developed in an unbiased way hence to find the track in 3D.
27250
27300
27350
27400
27450
27500
27550
27600
27650
27700
27750
6 8 10 12 14 16 18 20 22
Pixel 250
arb
Time (in mins)
0.000
0.002
0.004
0.006
0.008
0.010
0.012
0.014
6 8 10 12 14 16 18 20
GST
arb
Time (in mins)
• Based on CMSTR – track finder for CMS and work by David Colling and JH : Signal bands move with known velocity function.
• GST takes all pixels and for each time frame and each pixel calculates a velocity which a biomolecule would have to have to reach that pixel at that time.
• This is then histogrammed according to a weight determined by the Beer-Lambert absorption
• This results in an unbiased determination of velocity while retaining all peak shape information
• It requires a ‘virtual vertex’ to be assumed. Equiphase vertexing is a more specific application of GST
• Increases S/N & Retains Peak Shape
Generalised Separation Transform (GST)
Early use of vertexing to identify small signals
27250
27300
27350
27400
27450
27500
27550
27600
27650
27700
27750
6 8 10 12 14 16 18 20 22
Pixel 250
arb
Time (in mins)
Peak finding
ConstructEquiphase map
• Based on GST, TASSO/Aleph/Babar/D0 vertexing work and work by D. Sideris. The first stage of EVA processing is to perform peak finding on each of the 512 Electropherograms
• The time and amplitude of each peak is determined
• The Equiphase Map is constructed using every identified peak for each pixel
Equiphase Vertexing Algorithm (EVA)
Sample-run profile(Bacterial cell analysis)
The Vertex
Finding the vertex allows us to identify the unlabelled protein or DNA bands effectively increasing the signal to noise
Single vertexes are generally used but multiple vertexing can also be achieved
Detecting multiple vertices allows higher throughput, improved systematics and allows sample injected at different times to be identified e.g Virtual colour in DNA sequencingWork by Gary Taylor, Phil Lewis, John Hassard and others
Powerful Multipixel Detection Algorithms
Raw Data GST
EVA
Multiple Analyses
Overlay of the EVA processed data of nine consecutive E. coli lysate runs all separated under the same conditions.
Conventional PAGE
Conventional CE
Relative Standard DeviationPeak Time 0.98%Peak Height 4.56%
deltaDOT Peregrine
E.coli Analysis
Glycoproteins
Ribonuclease B Glycoforms
Relative Standard DeviationPeak Time 0.23%Peak Height 3.03%
Analysis of Antibody Standard
An overlay of EVA-processed data of 8 consecutive runs of Beckman Control Standard IgG in denaturing condition.
• We have established LFII as the most promising new approach in biotechnology. At the heart of the world’s biggest rapid vaccine development program
• LFII rewrites and reduces the compromises inherent in separation technologies, using a multisensor approach derived from other fields. Analysis power is a combination of several parameters, and LFII optimises this.
• LFII is agnostic as to target, and allows powerful relatively bias-free approaches to analysis
• LFII is based entirely on things particle physicists take for granted
• Particle physics cannot take for granted the continued goodwill and funding from governments without a redoubling of effort to show wealth creation.